Angiogenetic protooncogene ets-1 induced neovascularization is involved in the metastatic process of testicular germ cell tumors.

Abstract

Induction of angiogenesis is essential for tumor growth and metastasis. The role of angiogenetic factors and corresponding microvessel density in the development of metastasis of nonseminomatous testicular germ cell tumors (NSGCTs) is not clearly defined. Aim of the study was to gain new insights in the expression of the above described factors in different histological subtypes of metastatic and non-metastatic NSGCTs. Paraffin-embedded tissues of 39 NSGCTs (19 organ confined NSGCTs, pathological stage I and 20 NSGCTs with metastasis, pathological stage II) were immunohistochemical stained with antibodies for vascular endothelial growth factor (VEGF) with the ligands flt and flk, for the angiogenetic factor ets-1 as well as for endothelial markers CD34 and CD105. Areas representative of the invasive tumor were selected and the different histological subtypes were microdissected. For each subtype immunohistochemical expression in metastatic and organ confined tumors was assessed. Additionally, neovascularization was investigated by microvessel density and correlated to the markers of angiogenesis. VEGF expression was most often seen in teratoma components. There was no significant difference in the expression of VEGF, flt or flk in metastatic vs. organ confined tumors and their different histological subtypes seen. Only frequency and intensity for ets-1 expression differed significantly between metastatic and non-metastatic tumors. Using CD105 a significantly higher microvessel density was observed in stage II tumors. In addition, microvessel density determined by CD105 correlated significantly with ets-1 expression. VEGF and its receptors flt and flk seem not be involved in the progress of metastatic development of NSGCTs. Only expression of ets-1, a protooncogene involved in tumor angiogenesis, was significantly higher and more frequently seen in metastatic NSGCTs. A difference in microvessel density between tumor stages could only be observed if the specific CD105 antibody was used. In contrast, no difference in microvessel distribution between histological subtypes or tumor stages was observed if the pan-endothelial marker CD34 was used. Furthermore, ets-1 expression was significantly associated with CD105 microvessel density. Conclusively, ets-1 together with microvessel density determined by CD105 may have prognostic value in the multistep event of carcinogenesis.